tslib consists of the library libts and tools that help you calibrate and use it in your environment. There's a short introductory presentation from 2017.
tslib should be usable on various operating systems, including GNU/Linux, Freebsd or Android Linux. Apart from building the latest tarball release by ./configure, make and make install, tslib is available from the following distributors and their package management:
- Arch Linux and Arch Linux ARM -
pacman -S tslib - Buildroot -
BR2_PACKAGE_TSLIB=y - (Debian: Looking for a sponsor to include it)
TSLIB_TSDEVICE TS device file name.
Default (inputapi): /dev/input/ts
/dev/input/touchscreen
/dev/input/event0
Default (non inputapi): /dev/touchscreen/ucb1x00
TSLIB_CALIBFILE Calibration file.
Default: ${sysconfdir}/pointercal
TSLIB_CONFFILE Config file.
Default: ${sysconfdir}/ts.conf
TSLIB_PLUGINDIR Plugin directory.
Default: ${datadir}/plugins
TSLIB_CONSOLEDEVICE Console device.
Default: /dev/tty
TSLIB_FBDEVICE Framebuffer device.
Default: /dev/fb0
- On Debian,
TSLIB_PLUGINDIRprobably is for example/usr/lib/x86_64-linux-gnu/ts0. - Find your
/dev/input/eventXtouchscreen's event file and either- Symlink
ln -s /dev/input/eventX /dev/input/tsor export TSLIB_TSDEVICE /dev/input/eventX
- Symlink
- If you are not using
/dev/fb0, be sure to setTSLIB_FBDEVICE
This is just an example /etc/ts.conf file. Touch samples flow from top to bottom. Each line specifies one module and it's parameters. Modules are processed in order. Use one module_raw that accesses your device, followed by any combination of filter modules.
module_raw input
module median depth=3
module dejitter delta=100
module linear
see the below for available filters and their parameters.
With this configuration file, we end up with the following data flow through the library:
driver --> raw read --> median --> dejitter --> linear --> application
module module module module
Calibration is done by the linear plugin, which uses it's own config file /etc/pointercal. Don't edit this file manually. It is created by the ts_calibrate program:
# ts_calibrate
The calibration procedure simply requires you to touch the cross on screen, where is appears, as accurate as possible.
You may quickly test the touch behaviour that results from the configured filters, using ts_test_mt:
# ts_test_mt
One way to provide your resulting input behaviour to your system, is to use tslib's userspace input driver ts_uinput:
# ts_uinput -d
-d make the program return and run as a daemon in the background. Inside of /dev/input/ there now is a new input event device, which provides your configured input. You can even use a script like tools/ts_uinput_start.sh to start the ts_uinput daemon and create a defined /dev/input/ts_uinput symlink.
Remember to set your environment and configuration for ts_uinput, just like you did for ts_calibrate or ts_test_mt.
Let's recap the data flow here:
driver --> raw read --> filter --> (...) --> ts_uinput --> evdev read
module module module(s) application application
Variance filter. Tries to do it's best in order to filter out random noise coming from touchscreen ADC's. This is achieved by limiting the sample movement speed to some value (e.g. the pen is not supposed to move quicker than some threshold).
This is a 'greedy' filter, e.g. it gives less samples on output than receives on input. It can cause problems on capacitive touchscreens that already apply such a filter.
There is no multitouch support for this filter (yet). ts_read_mt() will
only read one slot, when this filter is used. You can try using the median
filter instead.
-
deltaSet the squared distance in touchscreen units between previous and current pen position (e.g. (X2-X1)^2 + (Y2-Y1)^2). This defines the criteria for determining whenever two samples are 'near' or 'far' to each other.
Now if the distance between previous and current sample is 'far', the sample is marked as 'potential noise'. This doesn't mean yet that it will be discarded; if the next reading will be close to it, this will be considered just a regular 'quick motion' event, and it will sneak to the next layer. Also, if the sample after the 'potential noise' is 'far' from both previously discussed samples, this is also considered a 'quick motion' event and the sample sneaks into the output stream.
Removes jitter on the X and Y co-ordinates. This is achieved by applying a weighted smoothing filter. The latest samples have most weight; earlier samples have less weight. This allows to achieve 1:1 input->output rate. See Wikipedia for some general theory.
-
deltaSquared distance between two samples ((X2-X1)^2 + (Y2-Y1)^2) that defines the 'quick motion' threshold. If the pen moves quick, it is not feasible to smooth pen motion, besides quick motion is not precise anyway; so if quick motion is detected the module just discards the backlog and simply copies input to output.
Linear scaling module, primerily used for conversion of touch screen
co-ordinates to screen co-ordinates. It applies the corrections as recorded
and saved by the ts_calibrate tool.
-
xyswapinterchange the X and Y co-ordinates -- no longer used or needed if the linear calibration utility
ts_calibrateis used. -
pressure_offsetoffset applied to the pressure value
-
pressure_mulfactor to multiply the pressure value with
-
pressure_divvalue to divide the pressure value by
Pressure threshold filter. Given a release is always pressure 0 and a press is always >= 1, this discards samples below / above the specified pressure threshold.
-
pminMinimum pressure value for a sample to be valid.
-
pmaxMaximum pressure value for a sample to be valid.
Simple debounce mechanism that drops input events for the specified time after a touch gesture stopped. Wikipedia has more theory.
-
drop_thresholddrop events up to this number of milliseconds after the last release event.
Skip nhead samples after press and ntail samples before release. This should help if for the device the first or last samples are unreliable.
Parameters:
-
nheadNumber of events to drop after pressure
-
ntailNumber of events to drop before release
Similar to what the variance filter does, the median filter suppresses spikes in the gesture. For some theory, see Wikipedia
Parameters:
-
depthNumber of samples to apply the median filter to
Check out the tests directory for examples how to use it.
ts_open()
ts_config()
ts_setup()
ts_close()
ts_reconfig()
ts_option()
ts_fd()
ts_load_module()
ts_read()
ts_read_raw()
ts_read_mt()
ts_reat_raw_mt()
The API is documented in our man pages. Possibly there will be distributors who provide them online, like Debian had done for tslib-1.0. As soon as there are up-to-date html pages hosted somewhere, we'll link the functions above to it.
Wikipedia has background information.
Usually, and everytime until now, libts does not break the ABI and your application can continue using libts after upgrading. Specifically this is indicated by the libts library version's major number, which should always stay 0. According to our versioning scheme, the major number is incremented only if we break backwards compatibility. The second or third minor version will increase with releases.
Officially, a tslib tarball version number doesn't tell you anything about it's backwards compatibility.
- libc (with libdl if you build it dynamically linked)
- libevdev - access wrapper for event device access, uinput too ("Linux API")
- libinput - handle input devices for Wayland (uses libevdev)
- xf86-input-evdev - evdev plugin for X (uses libevdev)
- ts_uinput - userspace event device driver for the tslib-filtered samples
- xf86-input-tslib - outdated direct tslib input plugin for X
- qtslib - outdated Qt5 qtbase tslib plugin
This is a complete example program, similar to ts_print_mt.c:
#include <stdio.h>
#include <stdlib.h>
#include <fcntl.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
#include <sys/time.h>
#include <getopt.h>
#include <errno.h>
#include <unistd.h>
#ifdef __FreeBSD__
#include <dev/evdev/input.h>
#else
#include <linux/input.h>
#endif
#include "tslib.h"
int main(int argc, char **argv)
{
struct tsdev *ts;
char *tsdevice = NULL;
struct ts_sample_mt **samp_mt = NULL;
struct input_absinfo slot;
unsigned short max_slots = 1;
int ret, i;
ts = ts_setup(tsdevice, 0);
if (!ts) {
perror("ts_setup");
return errno;
}
if (ioctl(ts_fd(ts), EVIOCGABS(ABS_MT_SLOT), &slot) < 0) {
perror("ioctl EVIOGABS");
ts_close(ts);
return errno;
}
max_slots = slot.maximum + 1 - slot.minimum;
samp_mt = malloc(sizeof(struct ts_sample_mt *));
if (!samp_mt) {
ts_close(ts);
return -ENOMEM;
}
samp_mt[0] = calloc(max_slots, sizeof(struct ts_sample_mt));
if (!samp_mt[0]) {
free(samp_mt);
ts_close(ts);
return -ENOMEM;
}
while (1) {
ret = ts_read_mt(ts, samp_mt, max_slots, 1);
if (ret < 0) {
perror("ts_read_mt");
ts_close(ts);
exit(1);
}
if (ret != 1)
continue;
for (i = 0; i < max_slots; i++) {
if (samp_mt[0][i].valid != 1)
continue;
printf("%ld.%06ld: (slot %d) %6d %6d %6d\n",
samp_mt[0][i].tv.tv_sec,
samp_mt[0][i].tv.tv_usec,
samp_mt[0][i].slot,
samp_mt[0][i].x,
samp_mt[0][i].y,
samp_mt[0][i].pressure);
}
}
ts_close(ts);
}
| Name | Introduced |
|---|---|
ts_close |
1.0 |
ts_config |
1.0 |
ts_reconfig |
1.3 |
ts_setup |
1.4 |
ts_error_fn |
1.0 |
ts_fd |
1.0 |
ts_load_module |
1.0 |
ts_open |
1.0 |
ts_option |
1.1 |
ts_read |
1.0 |
ts_read_mt |
1.3 |
ts_read_raw |
1.0 |
ts_read_raw_mt |
1.3 |
tslib_parse_vars |
1.0 |
Ideally you fork the project on github, make your changes and create a pull request. You can, however, send your patch to the current maintainer, Martin Kepplinger too. We don't have a mailing list at the moment. Before you send changes, it would be awesome if you checked the following:
- update the NEWS file's changelog if you added a feature
- update or add man pages if applicable
- update the README if applicable
- add a line containing
Fixes #XXin your git commit message; XX being the github issue's number if you fix an issue - update the wiki
For those creating tslib modules, it is important to note a couple things with regard to handling of the ability for a user to request more than one ts event at a time. The first thing to note is that the lower layers may send up less events than the user requested, because some events may be filtered out by intermediate layers. Next, your module should send up just as many events as the user requested in nr. If your module is one that consumes events, such as variance, then you loop on the read from the lower layers, and only send the events up when
- you have the number of events requested by the user, or
- one of the events from the lower layers was a pen release.
We loosely tie to the Linux coding style
A release can be done when either
- an issue tagged as "improvement" is implemented, or
- the previous release contains a user visible regression
The procedure looks like this:
- update the NEWS file with the changelog
- switch to a new release/ branch
- update configure.ac libts library versions
AC_INIT- includes the tslib package version. generally we increment the minor versionLT_CURRENT- increment only if there are API changes (additions / removals / changes)LT_REVISION- increment if anything changed. but ifLT_CURRENTwas incremented, set to 0!LT_AGE- increment only ifLT_CURRENTwas incremented and these API changes are backwards compatible (should always be the case, so it should matchLT_CURRENT)LT_RELEASE- matches theAC_INITpackage version
- update gitignore for autobuilt files
- autobuild and add the files for the tarball to git
- commit "add generated files for X.X release"
- git tag -s X.X
- git push origin release/X.X --tags
- make dist
- shas256sum > .sha256 for the 3 tarballs
- create a github release off the signed tag
- add the 6 files and the release notes
- publish and inform distributors
- celebrate!